There has been considerable debate about whether the Atlantic northern bluefin tuna exist as a single panmictic unit. We have addressed this issue by examining both mitochondrial DNA control region nucleotide sequences and nuclear gene ldhA allele frequencies in replicate size or year class samples of northern bluefin tuna from the Mediterranean Sea and the northwestern Atlantic Ocean. Pairwise comparisons of multiple year class samples from the 2 regions provided no evidence for population subdivision. Similarly, analyses of molecular variance of both mitochondrial and ldhA data revealed no significant differences among or between samples from the 2 regions. These results demonstrate the importance of analyzing multiple year classes and large sample sizes to obtain accurate estimates when using allele frequencies to characterize a population. It is important to note that the absence of genetic evidence for population substructure does not unilaterally constitute evidence of a single panmictic population, as genetic differentiation can be prevented by large population sizes and by migration.

In recent years, there has been considerable debate about whether the Atlantic northern bluefin tuna, Thunnus thynnus thynnus, reproduces as a panmictic unit. To address this question, we examined both mitochondrial DNA control region nucleotide sequences and nuclear gene ldhA allele frequencies in replicate samples of northern bluefin tuna from the Mediterranean Sea and the northwestern Atlantic Ocean. AMOVA analyses of both types of data revealed no significant differences between samples from the two regions. These results demonstrate the importance of analyzing multiple year classes and large sample sizes in stock structure analyses. In addition, larval samples from the Gulf of Mexico and the Mediterranean Sea were not significantly different from each other or from the other samples when control region sequences were compared. However, despite the strong evidence presented here, failure to find genetic evidence for population substructure does not constitute evidence for a single panmictic population. It is possible that multiple subpopulations do exist, and that genetic differentiation at the loci analyzed in this study has not occurred because of large population sizes and/or low levels of reproductively successful migration between the subpopulations. RÉSUMÉ

In recent years, there has been considerable debate about whether the Atlantic northern bluefin tuna, Thunnus thynnus thynnus, reproduces as a panmictic unit. To address this question, we examined both mitochondrial DNA control region nucleotide sequences and nuclear gene ldhA allele frequencies in replicate samples of northern bluefin tuna from the Mediterranean Sea and the northwestern Atlantic Ocean. AMOVA analyses of both types of data revealed no significant differences between samples from the two regions. These results demonstrate the importance of analyzing multiple year classes and large sample sizes in stock structure analyses. In addition, larval samples from the Gulf of Mexico and the Mediterranean Sea were not significantly different from each other or from the other samples when control region sequences were compared. However, despite the strong evidence presented here, failure to find genetic evidence for population substructure does not constitute evidence for a single panmictic population. It is possible that multiple subpopulations do exist, and that genetic differentiation at the loci analyzed in this study has not occurred because of large population sizes and/or low levels of reproductively successful migration between the subpopulations.

SUMMARY In recent years, there has been considerable debate about whether the Atlantic northern bluefin tuna (Thunnus thynnus) exist as a panmictic unit. This question is addressed by examining both mitochondrial DNA control region nucleotide sequences and nuclear gene ldhA allele frequencies in replicate samples of northern bluefin tuna from the Mediterranean Sea and the northwestern Atlantic Ocean. Preliminary analyses of both the ldhA and the mtDNA data had provided tentative evidence for population subdivision of Atlantic northern bluefin tuna. However, with increased sample sizes, AMOVA analyses of both types of data revealed no significant differences between samples from the two regions. Similarly, pairwise comparisons of multiple year class samples from the two regions provided no evidence for population subdivision. These results demonstrate the importance of analyzing multiple year classes and large sample sizes in stock structure analyses. Also, it is important to note that failure to find genetic evidence for population substructure does not constitute evidence for a single panmictic population. Multiple subpopulations could exist, and genetic differentiation could be prevented by large population sizes and/or low levels of migration. RESUME